| Author | Message | |||
     Matthew Lloyd Davies (mdavies) New member Username: mdavies Post Number: 1 Registered: 10-2004 |
I have been scratching my head a little over this. I am trying to model the effects of fire on a box of fireworks. I have begun with a 2D problem and manually set up the boundaries for two basic fireworks. The first firework is initiated and it starts to burn. Heat conduction through the wall causes the second to ignite. I would like to move to n fireworks but I don't want to have to define the boundary of each individually. Is it possible to set up an array and associated script to generate the boundaries and regions of n fireworks? Eventually I would like to move to an array of 3D cylinders. Can the same principles be applied in that case. I have attached my work so far. Thank you in advance for your help.
| |||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 232 Registered: 06-2003 |
The example problem "Samples | Misc | arrayrepeat.pde" shows the use of ARRAYS and REPEAT loops to build an array of objects. It sounds like this is what you want to do. You can apply repetition in a 2D figure and specify as many layers of this pattern as you want in 3D. Incidentally, you have specified FIXDT and a timestep of 0.5. This is much too large a timestep, and drives the problem into catastrophe. With the timestep control enabled, FlexPDE seems content with a timestep of about 0.01(PREFER_SPEED) or 0.003(PREFER_STABILITY) after the initial ignition flux has passed. With a fixed timestep, you can get away with 0.02. See attached.  | |||
| Matthew Lloyd Davies (mdavies) New member Username: mdavies Post Number: 2 Registered: 10-2004 |
Robert, Thank you for pointing that out. I was messing around seeing how things worked in FlexPDE and obviously forgot to change it back. PS I've not been using Flex for long but I've picked up the basics pretty quickly. We're looking to upgrade to the full version soon. It's a great piece of software which we're hoping to use extensively over the next few years to investigate the hazards associated with the transport and storage of fireworks. We would gladly send you reprints of publications which use FlexPDE. | |||
| Matthew Lloyd Davies (mdavies) Junior Member Username: mdavies Post Number: 3 Registered: 10-2004 |
I have updated my code to include a repeat loop to generate n fireworks. I have also used a repeat loop to create n plots of temperature along the central axis of each. Is it possible to re-write this code such that each curve is on the same chart? Secondly I don't seem to be able to figure out a way to use a repeat loop to generate n history plots (one at the centre of each article). From the attached code you can see that I have had to manually code in each point, which means changing this section of the code whenever I change n. Any ideas would be appreciated. ty
| |||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 233 Registered: 06-2003 |
Each plot has a unique physical position, so there is no way to overlay plots along different baselines. Sorry. The repeat issue is a shortcoming of the way we wrote the FlexPDE script grammar. There was no provision for repeats in histories. We will correct this in version 4.2.1, but for the time being, you will have to write the positions manually. We are always eager to receive reprints of articles using FlexPDE. As long as there are no copyright problems, we would like to post them on our Reprints web page. If you are working with a student version and intend to upgrade to the full version, you should request a trial license, so you can work in full capability mode. Send the Computer ID from the Help|Register dialog to sales@pdesolutions.com with a request for a trial license. | |||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 234 Registered: 06-2003 |
We have released FlexPDE version 4.2.1. This version allows repeats in history statements, either surrounding the history plot itself, or around the "at" points. This version also corrects an error in writing TABLE files from irregular domains, in which some of the table files fall outside the domain. |
Pyrotechnic array